Architecture and Neurocytology of Monkey Cingulate Gyrus

Home , Central sulcus, Cingulate cortex, Cingulate sulcus, Gyrus, Posterior cortex, Supplementary motor area

THE JOURNAL OF COMPARATIVE NEUROLOGY 485:218–239 (2005)

BRENT A. VOGT,1,2* LESLIE VOGT,1,2 NURI B. FARBER,3 AND GEORGE BUSH4,5 1Cingulum NeuroSciences Institute, Manlius, New York 13104 2State University of New York Upstate Medical University, Syracuse, New York 13210 3Department of Psychiatry, Washington University, St. Louis, Missouri 63110 4Department of Psychiatry, Harvard Medical School, Boston, Massachusetts 02115 5Massachusetts General Hospital, Charlestown, Massachusetts 02129

ABSTRACT Human functional imaging and neurocytology have produced important revisions to the organization of the cingulate gyrus and demonstrate four structure/function regions: anterior, midcingulate (MCC), posterior (PCC), and retrosplenial. This study evaluates the brain of a rhesus and 11 cynomolgus monkeys with Nissl staining and immunohistochemistry for neuron-specific nuclear binding protein, intermediate neurofilament proteins, and parvalbumin. The MCC region was identified along with its two subdivisions (a24Ј and p24Ј). The transition between areas 24 and 23 does not involve a simple increase in the number of neurons in layer IV but includes an increase in neuron density in layer Va of p24Ј,a dysgranular layer IV in area 23d, granular area 23, with a neuron-dense layer Va and area 31. Each area on the dorsal bank of the cingulate gyrus has an extension around the fundus of the cingulate sulcus (f 24c, f 24cЈ, f 24d, f 23c), whereas most cortex on the dorsal bank is composed of frontal motor areas. The PCC is composed of a dysgranular area 23d, area 23c in the caudal cingulate sulcus, a dorsal cingulate gyral area 23a/b, and a ventral area 23a/b. Finally, a dysgranular transition zone includes both area 23d and retrosplenial area 30. The distribution of areas was plotted onto flat maps to show the extent of each and their relationships to the vertical plane at the anterior commissure, corpus callosum, and cingulate sulcus. This major revision of the architectural organization of monkey cingulate cortex provides a new context for connection studies and for devising models of neuron diseases. J. Comp. Neurol. 485:218–239, 2005. © 2005 Wiley-Liss, Inc.

Indexing terms: cingulate cortex; neuroﬁlament proteins; cingulate motor areas; midcingulate cortex; retrosplenial cortex; pyramidal neurons

The primate cingulate gyrus was early considered to be perigenual anterior cingulate cortex (ACC), midcingulate a uniform structure with a role in limbic function (Broca, cortex (MCC), posterior cingulate cortex (PCC), and retro- 1878; Papez, 1937; MacLean, 1990). Although it has also splenial cortex (RSC). been recognized to have a dual structure with different A key aspect of the four-region model is division of connections (Baleydier and Mauguiere, 1980; Vogt et al., Brodmann’s (1909) ACC into a perigenual and midcingu- 1987) and with an executive function in the anterior and late regions and this differentiation is pivotal to under- evaluative role for the posterior parts (Vogt et al., 1992), many cytoarchitectural studies showed the human cingulate gyrus to be more complex than the dual model sug- Grant sponsor: National Institutes of Health; Grant number: NINDS gests (Smith, 1907; Brodmann, 1909; Vogt and Vogt, 1919; RO144222; Grant number: NIAAA AA06916 (to B.A.V.); Grant number: von Economo and Koskinas, 1925; Rose, 1927) and numer- NIA PPG11355 (to N.B.F., B.A.V.); Grant sponsor: National Science Foun- ous human functional imaging studies show that cingu- dation; Grant number: 242229 (to G.B). late cortex is involved in more than just two essential *Correspondence to: Brent A. Vogt, Cingulum NeuroSciences Institute, 4435 Stephanie Drive, Manlius, NY 13104. E-mail: [email protected] functions. Structure/function correlations with human Received 12 October 2004; Revised 13 December 2004; Accepted 13 neurocytology and imaging, electrical stimulation, and December 2004 stroke ﬁndings suggest there are four rather than just two DOI 10.1002/cne.20512 regions (Vogt et al., 1997, 2004). The four regions are Published online in Wiley InterScience (www.interscience.wiley.com).

© 2005 WILEY-LISS, INC. MONKEY CINGULATE GYRUS 219 standing the functional organization of primate cingulate large neurons in layers III and V that appears to parallel cortex. The ACC has reciprocal connections with the Braak’s primitive gigantopyramidal field, and they iden- amygdala, projections to the nucleus of the solitary tract tified a rostral extension termed cmc2 with smaller, al- and dorsal motor nucleus of the vagus, regulates auto- though still large, neurons. This latter area might regu- nomic output, and stores emotional memories (Neafsey et late muscles in the hindlimb, lower trunk, and tail al., 1993; Vogt et al., 1997, 2003, 2004). In contrast, the (Rizzolatti et al., 1996) and may have the highest density MCC regulates skeletomotor function through its projec- of cingulospinal projection neurons (Dum and Strick, tions to the spinal cord, receives only modest amygdala 1991), whereas cmc1 may regulate forelimb, upper trunk, input to its anterior part and does not directly regulate and neck muscles (Rizzolatti et al., 1996). The cmc dichot- autonomic outputs, and has a prominent projection from omy has not been considered in the monkey nor have posterior parietal cortex (Vogt et al., 1997, 2004). The transitions from cingulate to premotor areas in the cingu- posterior part of human MCC contains the primitive gi- late sulcus dorsal bank. gantopyramidal motor field of Braak (1976), which in Second, a previous map of monkey cingulate cortex monkey is termed area 24d or the caudal cingulate motor shows a simple transition from agranular area 24 to gran- area (cCMA). The cCMA has somatotopically organized ular area 23; presumably due to a buildup in layer IV skeletomotor projections to the spinal cord (Biber et al., neurons similar to that shown by Brodmann (Vogt et al., 1978; Dum and Strick, 1991, 1993, Matelli et al., 1991; 1987; Morecraft et al., 2004). Recent human studies show Luppino et al., 1991) and large pyramids in layer V that a more complicated transition pattern, including differen- express intermediate neurofilament proteins (Nimchinsky tiation of layer Va in midcingulate cortex (Vogt et al., et al., 1996, 1997) and neurons that project to motor cortex 1995, 2003, 2004), which includes a neuron-dense layer Va (Morecraft and Van Hoesen, 1992; Nimchinsky et al., in pMCC, a dysgranular division of area 23 (area 23d), and 1996). The rostral cingulate sulcus has the rostral cingu- a dorsal posterior cingulate cortex (d23a/b) with granular late motor area (rCMA), which differs from the cCMA layers IV and Va. This sequence of transitional events because the former has longer premovement activity on- needs to be evaluated in the monkey. It is also unclear set, responses associated with the changing reward fea- whether area 32 extends dorsal to the rostral CMA as it tures of a movement, and projections to the presupple- does in human. Thus, a complete analysis of MCC as a mentary motor part of the striatum rather than to that functional and cytological entity and its adjacent areas is part which receives primary motor cortex input and dif- needed in the monkey. ferences in corticocortical connections (Shima et al., 1991; Third, the monkey PCC has been differentiated into two Luppino et al., 1991; Van Hoesen et al., 1993; Rizzolatti et parts on the basis of thalamic (Shibata and Yukie, 2003), al., 1996; Shima and Tanji, 1998; Takada et al., 2001; prefrontal (Vogt and Barbas, 1988), and superior temporal Akkal et al., 2002). Finally, the duality of the human MCC (Yukie, 1995) connections. These two divisions also have has been established with anterior (aMCC) and posterior been immunohistochemically and functionally defined in (pMCC) parts based on the two CMAs and different cyto- human (Vogt et al., personal communication). It is not architectures (Vogt et al., 2003). known to what extent the monkey PCC has dorsal (dPCC Progress over the past decade in human neuroscience or area d23a/b) and ventral vPCC or area v23a/b) divi- has raised many questions about the organization of the Ј Ј primate cingulate gyrus, particularly in the monkey, and sions. Where is the junction between areas p24a /b and these questions frame the present research. First, Zilles et area 23a/b and between areas 24d and 23c in the caudal al. (1996) showed a caudal area cmc1 in human with very cingulate sulcus? Also, what are the characteristics of the border between pMCC and PCC, and is there a cytological basis for dorsal and ventral area 23a/b? Fourth, from a methodological perspective, it should be noted that cytoarchitectural studies often use low magni- Abbreviations fication optics to assess Nissl preparations and often focus ac anterior commissure on only a part of the gyrus. Cellular resolution using ACC anterior cingulate cortex neuron-specific nuclear binding protein (NeuN) has the cas callosal sulcus benefit of not staining glial or vascular cells and enhances ces central sulcus cgs cingulate sulcus laminar differences not apparent with Nissl stains. Immu- cml caudomedial lobule; mainly area v23b nohistochemistry can be used to assess particular proteins rCMA, cCMA cingulate motor areas, rostral or caudal expressed by specific neurons as with antibodies to inter- gcc genu of the corpus callosum mediate nonphosphorylated neurofilament proteins (NFP; ir immunoreactivity MCC midcingulate cortex SMI32 antibody) for large pyramidal neurons and mr marginal ramus calcium-binding proteins for local-circuit interneurons. NeuN, N neuron-specific nuclear binding protein Dombrowski et al. (2001) evaluated different classes of NFP neurofilament proteins PCC posterior cingulate cortex calcium-binding protein-expressing neurons in monkey Parv, P parvalbumin frontal lobe and concluded that parvalbumin was the most pc posterior commissure effective in specifying structural differences among areas. pos parieto-occipital sulcus Parvalbumin-immunoreactive neurons are in all layers, rs rostral sulcus RSC retrosplenial cortex they are an interneuron marker that complements the scc splenium of the corpus callosum large neuron labeling of SMI32, and it is valuable for SEM standard error of the mean distinguishing areas on the cingulate gyral surface and SMI32, S antibody to nonphosphorylated intermediate neurofila- dorsal bank (Nimchinsky et al., 1997). Because different ments spls splenial sulcus laminar patterns of neuronal markers make some mark- VCA vertical plane at the anterior commissure ers better to identify individual areas (Carmichael and 220 B.A. VOGT ET AL.

Price, 1994), we added parvalbumin to assess cingulate overnight at 4°C. After the sections were incubated in the neurocytology. primary antibody, they were then rinsed in PBS and in- The strategy for this work involved three stages. First, cubated in biotinylated secondary antibody at 1:200 in a re-analysis of an original Nissl-stained, rhesus monkey PBS/Triton-X/BSA for 1 hour. After rinses in PBS, sec- case was performed to assess the midcingulate concept in tions were incubated in ABC solution (1:4; Vector Labora- monkey with classic methods, and a re-analysis considers tories, Burlingame, CA) in PBS/Triton-X/BSA for 1 hour the border and transition between agranular area 24 and followed by PBS rinses and incubation in 0.05% diamino- granular area 23. Second, immunohistochemistry for benzidine, 0.01% H2O2 in a 1:10 dilution of PBS for 5 NeuN, NFP, and parvalbumin macrophotography was minutes. After final rinses in PBS, sections were mounted, used to evaluate each of the above questions raised from air-dried, counterstained with thionin, dehydrated, and the perspective of human cortex and high-magnification cover-slipped. Before a final series was prepared, two to neurocytology analyzed to determine the features of each three sections were processed from at least three blocks area. Third, cases were flat mapped with reference to the distributed over the cingulate gyrus according to optimal vertical plane at the anterior commissure (VCA) to coreg- reaction conditions to check the reaction before a complete ister these and other findings to resolve questions of struc- series was prepared to ensure that all reagents and tis- ture and function correlations. sues were optimal. Each new batch of antibody was tested for specificity by excluding the primary antibody from the reaction, including the peptide blocker parvalbumin in the MATERIALS AND METHODS reaction, and all sections were counterstained to ensure that nuclei/layers with high immunoreactivity were so Subjects and tissue processing stained. Once the sections were mounted, they were either One adult, male rhesus monkey was used to initiate dried and cover-slipped or first counterstained with thi- these studies by serving as a point of re-analysis from onin for 3 min (0.05% thionin, 3.7% sodium acetate, 3.5% Vogt et al. (1987). The case was embedded in celloidin, and glacial acetic acid, pH 4.5). This counterstaining of SMI32 the left hemisphere was cut coronally at a 30 ␮m thickness ensured accurate localization of impregnated neurons to into 285 sections and stained with cresyl violet. Once a particular layers in instances where only a few layers of survey was completed and documented in this case, 11 neurons were immunoreactive. adult cynomolgus monkeys were prepared. These animals (7 female, 4 male; weight, 4.02 Ϯ 1.06 kg) were anesthe- Analytical methods tized with ketamine (15 mg/kg, i.m.) and then with sodium The range of body weights for eight cynomolgus mon- pentobarbital (40 mg/kg, i.v.) according to a protocol ap- keys was 3–4.3 kg, and there was one outlier (i.e., 2 SD proved by the Committee for the Humane Use of Animals from mean) of 6.8 kg. Because variation in brain sizes at SUNY Upstate Medical University (Syracuse, NY), should determine the number and value of flat maps Wake Forest University School of Medicine (Winston- needed for this analysis, calibrated digital photographs of Salem, NC), and Washington University (St. Louis, MO). the medial surface were used to calculate the distance The animals were perfused intracardially in the following between the rostral tip of the genu and caudal tip of the sequence: (1) 1 ml of 0.1% sodium nitrite injected into the splenium of the corpora callosi. The range was 2.23–2.85 heart, (2) 300 ml 0.9% cold sodium chloride, (3) 1 liter of cm, and the mean Ϯ SEM was 2.55 Ϯ 0.217 cm. The 1% cold paraformaldehyde, (4) 1.5 liters of 4% cold para- largest animal in body weight had a distance of 2.8 cm, formaldehyde over a period of 40 minutes. Brains were and there were two other animals with measurements of removed, bisected, post-fixed in 4% paraformaldehyde for 2.8 and 2.85 cm. The correlation coefficient (r) of body 2–3 days in the refrigerator, and put through a graded weight and corpus callosum length for the 11 cases was series of 10%, 20%, and 30% sucrose until the brain sunk 0.06, showing no correlation. Thus, flat map variation is in each solution for refrigerator storage and cutting. dominated by measurement variation and the quality of These 11 animals were selected from a larger collection staining and border localization rather than differences in because they had full series through cingulate cortex with the size of the cingulate gyrus. NeuN and SMI32 immunohistochemistry, and three had Digital photographs of the medial surface were oriented an additional series with parvalbumin. Each case was parallel to the ac-pc line, and a perpendicular plane was oriented perpendicular to the anterior–posterior commis- defined at the anterior commissure (VCA). The medial sural line (ac-pc plane; Figs. 1, 2) and digitally photo- surface was flattened in one dimension only so that mea- graphed. Three coronal blocks were cut, and the brains surements from the vertical plane at the anterior commis- were rephotographed. All but one case was cut in the sure would be accurate in the rostrocaudal or “y” axis. coronal plane in a cryostat at a 40 ␮m thickness into six Maps were generated according to the following proce- series. One series each was prepared for NeuN (Chemicon, dure: (1) The digital image of the medial surface was Temecula, CA), SMI32 (Sternberger Monoclonal, Balti- rotated so the long axis of the corpus callosum was hori- more, MD) for NFP followed by counterstaining with thi- zontal. (2) The superior frontal and parietal sulci were onin, and parvalbumin (Chemicon) in the following man- removed and coronal cutting surface orientations marked. ner. Sections were pretreated with 75% methanol/25% (3) The dorsal edge of the corpus callosum was drawn and peroxidase, followed by a 3-minute pretreatment with for- repositioned above the photograph. (4) Matched sections mic acid (NeuN only) and then a washing with distilled of NeuN, SMI32, and parvalbumin were selected at equal water and two washes in phosphate buffered saline (PBS, distances for 19 levels, and each was digitally photo- pH 7.4). Sections were incubated in primary antibody in graphed at 1ϫ including a 1-mm calibration bar and PBS (dilutions: SMI32, 1:10,000, mouse; NeuN, 1:1,000, printed as triads. Each triad of photographs was printed mouse) and parvalbumin (1:3,000, mouse) containing 0.3% together on a single page. (5) Measurements were made at Triton X-100 and 0.5 mg/ml bovine serum albumin (BSA) mid-cortical levels of the distance between each gyral apex MONKEY CINGULATE GYRUS 221 and sulcal fundus, and dots were plotted as the distance Cynomolgus monkey: Surface features from the dorsal surface of the corpus callosum on the flat and flat map map. In two instances, the flat map was coregistered to surface renderings in other studies using different layers The medial surface is oriented parallel to the ac-pc in Photoshop. The coregistration was performed by align- plane in Figure 2 and the VCA drawn perpendicular ing the dorsal edge of the corpus callosum, anterior com- thereto. The plane of coronal sections for this case is missure, cingulate sulcus, and paracentral lobule in the shown with arrowheads at the face of each of the three flat map with those in each data set from Dum and Strick blocks. The corpus callosum was drawn at its dorsal bor- (1991) and Rizzolatti et al. (1996). der, placed above the medial surface photograph, and Once a survey series of low-magnification prints was flattening produced in one dimension by extension from the corpus callosum. The retrosplenial areas were drawn available, a centroid level was defined for each area (i.e., by extension below the dorsal part of the corpus callosum. approximate center in all planes of section) using the flat The splenial sulcus (spls) in this case is not directly at- map and higher magnification assessments and documen- tached to the cingulate sulcus (cgs), although there is a tation were made and the digital images were imported vascular indentation, and RSC (shaded area in Fig. 2) is into different palette layers in Adobe Photoshop 6.0. This placed ventral to the dorsal border of the callosum so that strategy was crucial, because the three immunohisto- the overlying area 23 can be represented. The knee of the chemical sections from the each area were temporarily cgs at the inflection of the mr is noted with an arrow in merged to ensure exact alignment of all cortical layers. Figure 1, because this is an important point of area con- This was accomplished by reducing the opacity of an over- fluence; just ventral to the knee area 23d on the gyral lying layer (SMI32 or parvalbumin) and coregistering the surface and area 23c in the cgs disappear. Finally, the sections based on pia matter, white matter, and large caudomedial lobule forms the terminal part of the poste- neurons in layers Va and IIIc (Figs. 9, 12). rior cingulate gyrus and lies just dorsal to the ac-pc line. Flat maps were generated like that shown in Figure 2 with reference to the genu of the corpus callosum, VCA, RESULTS knee of the cgs, and caudal edge of the splenium of the Rhesus monkey corpus callosum. To verify that the brains were a similar size, the corpora callosi were measured from the rostral The gyral surface of a Nissl-stained rhesus monkey case tip of the genu to the caudal tip of the splenium. This was evaluated to determine whether the characteristics of distance was 2.55 Ϯ 0.217 cm, and there was no correla- transition from area 24 to area 23 observed in human tion between body weight and the length of the corpus studies are applicable to the monkey. Of the 240 sections callosum. Thus, coordinates in the “y” and “z” planes for through the left cingulate gyrus, 30 were photographed at any area in this case are similar for all adult cynomolgus the “b” subdivisions to assess cytoarchitectural transition monkeys with less than a 10% measurement error. along the full length of the gyrus. An equally spaced Although the basic layout of cynomolgus cingulate cor- sample series of 10 sections are shown in Figure 1 where tex is similar to that reported for rhesus monkey, there magnifications were selected to emphasize mid-cortical have been several modifications based on rhesus monkey layers, where the transition from agranular to granular (Vogt, 1993; Vogt et al., 1997) and human observations cortex occurs, to identify the location of the first layer IV (Vogt et al., 2003, 2004) and these studies are reported neurons. The star at section 159 in Figure 1 refers to the herein. First, MCC is identified with a prime following original border between areas 24 and 23 (Vogt et al., area 24 to refer to its posterior placement and agranular 1987), and this border is 3 mm caudal to the VCA. The form, and it is composed of seven unique cytoarchitectures first evidence of a layer IV is in section 191 (arrow to layer and three fundal extensions. Second, transition to the IVd). At this level, there is a bridge of neurons between posterior cingulate region occurs by means of a dysgranu- layers IIIc and Va, and this lack of a complete layer IV is lar area 23d between areas 23 and p24Ј. Third, PCC is characteristic of a dysgranular layer IV. Because the pho- composed of dorsal and ventral divisions. tographs are aligned at the superficial border of layer Va, it can be seen in section 159 that layer Va has a neuron- Anterior cingulate cortex dense and granular appearance. This feature is character- The ACC includes areas 32 and 24a rostral to the genu, istic of pMCC and may have contributed over the past area 24b dorsal to and 25 ventral to the genu, and area 24c century to mislocalization of the border between areas 24 with its fundal extension (f 24c) in the cgs. Figure 3 shows and 23 when viewed with Nissl stains. No such neuron the most rostral part of ACC, including areas 24 and 32; density occurs in layer Va of areas 24b or a24bЈ. In addi- ventral area 9 (9v) is described below, because it is more tion, claims that area 24 is dysgranular seem to be incor- fully elaborated dorsal to the genu. Area 32 is a dysgranu- rect as can be seen in the high-magnification photographs lar cortex with a thin layer IV. Only layer Va neurons of Figure 1; layer IIIc and Va directly abut in areas p24bЈ, express NFPs, with very few such cells in layers IIIc and a24bЈ, and 24b and there is no dysgranular layer. Vb (Fig. 3). In contrast, area 24c has a broad layer of Thus, the monkey has an MCC with areas a24Ј and p24Ј NFP-expressing neurons in layer V, but activity in layer as in human, and there is a complex transition between IIIc is still weak. Generally, the large neurons in both areas 24 and 23, i.e., there is not a simple and progressive divisions of layer V are larger in area 24c than in area 32 increase in the number of layer IV neurons. Rather, there (Fig. 3, NeuN at high magnification). Layer VI in area 24c is an increase in small-neuron density of layer Va of p24Ј is also much more dense and differentiated than it is in when compared with a24Ј, a dysgranular layer IV in area area 32. 23d, and a granular layer IV in area 23 in conjunction with Subgenual cingulate cortex is composed of areas 24a, a dense layer Va. 25, and a frontal cortex extension termed area 10m (Car- 222 B.A. VOGT ET AL.

Fig. 1. Systematic sampling of cresyl violet-stained sections the dysgranular layer in area 23d (arrow, 191) and the dysgranular through a rhesus monkey case at two magnifications to assess archi- nature of this cortex is emphasized by the fact that layer IIIc/Va pyra- tecture in “b” areas along the cingulate gyrus. Layers in sections for mids interact at one point, whereas area 23b caudally has an unbroken this and subsequent figures are aligned on the border between layers layer IV. Higher magnification shows that layer Va neurons in sections IIIc and Va or layers IV and Va. The star on the surface photograph 98–130 are relatively large; in 140–159, large pyramids are somewhat is the position of the Brodmann border, and this “border” is 3 mm smaller; whereas in 169–180, there are many small neurons in layer Va caudal to the vertical plane at the anterior commissure (VCA). The giving the appearance of a granular layer; this is not layer IV, however. middle cortical layers are magnified to identify changes in small cgs, cingulate sulcus; mr, marginal ramus; pos, parieto-occipital sulcus; neuron densities in layers IV and Va. The first evidence of a layer IV is rs, rostral sulcus; spls, splenial sulci. Scale bars ϭ 100 ␮m.

michael and Price, 1994) as shown in Figure 4. Area 24a dense layer VI (NeuN) with many parvalbumin- surrounds the genu and has a broad and undifferentiated immunoreactive (-ir) neurons and dendrites. In contrast, layer V with substantial expression of NFPs. It also has a area 25 has much weaker NFP expression in layer V, MONKEY CINGULATE GYRUS 223

Fig. 2. Flattened reconstruction of the medial surface. The curved shading as is the RSC in the callosal sulcus (cas). Outlines of the arrows designate the knee of the cingulate sulcus (cgs) where it emits surface references are in black lines; each cytoarchitectural area is the marginal ramus (mr). Flattening was performed in one dimension outlined with a thick and less opaque line; and the lines are dashed parallel to the VCA, and measurements from the VCA (“y” plane) where subdivisions within an area are represented. Each division in reﬂect an accurate position, whereas measurements in the “z” and “x” the scale is 1 mm here and in subsequent medial surface photographs. planes are distorted. The fundal areas in the cgs are marked with For other abbreviations, see list. 224 B.A. VOGT ET AL.

Fig. 3. Location and composition of areas 32 and 24c in the rostral cgs. Higher magnification shows the dysgranular layer IV in area 32 with NeuN (N), the lack of one in area 24c, and the much higher level of SMI32 immunoreactivity (S) in area 24c than area 32. For abbreviations, see list. Scale bars ϭ 1mm in left column and middle column; 500 ␮m in right column. poorly differentiated layers V and VI, and fewer Cortex on the dorsal bank of the cgs is not “cingulate” in parvalbumin-expressing neurons in layer VI. Finally, area structure, although it has transitional features. The ven- 10m has a dense layer V, relatively few NFP- and tral area 9 is labeled 9v in Figure 5, and it has a slender parvalbumin-expressing neurons and axonal plexuses, layer Va, although it is less prominent than in area 24c. At and a dysgranular layer IV. Although there are some higher magnification, the larger layer Va neurons are parvalbumin-ir neurons in layer IV, no appreciable ex- apparent in area 9v. The cingulate “signature” of a very pression of NFPs occurs. neuron-dense layer Va that is disproportionate in relation Cortex in sulcal ACC is shown in Figure 5. Pivotal to to the less neuron-dense layer IIIc is not prominent in identifying individual layers in each cingulate area is area 9v. Importantly, area 9v has very large layer IIIc using designations that are consistent with those for neo- pyramids (Fig. 5, NeuN and SMI32; high-magnification cortex. Because cortical layers are differentially distorted photographs), which is not true for area 24c. Parvalbumin in the fundus with deep layers being proportionately more also distinguishes between these areas, because the layer stretched and superficial layers compressed, a level adja- VI expression in area 24c is very weak in area 9v, whereas cent to area 4 was selected so that Betz neurons in layer the number of labeled neuron and axons in superficial Vb could be identified in SMI32 of a medioventral division layers is much greater than in area 24c. Thus, although of area 4 (4mv; Fig. 5). The Betz neuron closest to fundal area 9v has transitional features characteristic for both area 24d (f24d) is circled, and the layers are identified in areas 24c and 9, it appears to share more features with the fundus and on the ventral bank of the cgs in area 24d. area 9, in particular, the fully developed and NFPϩ layer The most-dense plexus and somatic impregnation is in IIIc pyramidal neurons and very large pyramids in both layer Vb, whereas the neurofilament-sparse layer is in divisions of layer V. layer Va. Area 24c has the thickest and least-differentiated NFP- Cortex on the dorsal and ventral banks of expressing deep pyramidal layers V and VI of any part of the cingulate sulcus the cingulate gyrus. There are also very few layer IIIc neurons expressing NFPs, whereas f24c does have an Most of the cingulate gyrus abuts frontal cortex, except enhancement of them in layer IIIc. Layer Va is present for part of the PCC. In addition to defining the specific (Fig. 5; NeuN); however, it overlaps to some extent with organization of each cingulate area, an important ques- the NFP plexus, and the differentiation between layers Va tion is to what extent cortex on the dorsal bank of the and Vb is not pronounced. Parvalbumin expression shows sulcus is “cingulate” or “frontal” as partially considered two prominent layers of neuron labeling: one in layer Va above. This question is best resolved with low- and another in layer VI; the latter of which is more dense. magnification macrophotographs of the entire sulcus as in MONKEY CINGULATE GYRUS 225

Fig. 4. The ventral and rostral areas are differentiated with III, V, and VI (NeuN) and heavy NFP expression in layer V, and a NeuN, SMI32, and parvalbumin (Parv). A Parv plexus is particularly similar motif is observed in area 25 (SMI32), although NFP expres- notable in layer VI of area 24a and differentiation of areas 25 and 10m sion is much more limited in this latter area. For abbreviations, see is clear in all preparations. The composition of area 24a is shown at list. Scale bars ϭ 1 mm in left (top, bottom); 0.5 mm in right. higher magniﬁcation (arrow in Parv), with the homogeneous layers

Figure 6 and higher magniﬁcations to detail particular a24cЈ (fa24cЈ) abuts a ventral part of area 6a␤ (v6a␤), the neuron structures and protein expression patterns. Figure fundal division of area p24cЈ (fp24cЈ) abuts the ventral 6 shows three rostrocaudal levels of the cingulate sulcus division of area 6a␣ (v6a␣), whereas a medioventral divi- and shows that the dorsal bank is quite different in archi- sion of area 4 (4mv) abuts area 24d. Area 4mv is selected tecture than the ventral bank. The fundal part of area in Figure 6 to demonstrate how signiﬁcantly different the 226 B.A. VOGT ET AL.

Fig. 5. Anterior cingulate sulcal cortex is composed of area 24c, its greatest, an SMI32 section was used from a posterior level where area fundal extension (f24c), and area 9 in the dorsal bank. Although there 4mv abuts area 24d. The large Betz neurons in layer Vb are readily is a bilaminar pattern in parvalbumin expression in areas 9v, f24c, identified (one is circled), and this finding provides the gold standard and 24c, the level of staining in layer VI is much lower in 9v than area for subsequent labeling of layers. The three neuron markers and 24c, and the density of NFPs in layers IIIc and V is massive in merging the images for them ensures accurate laminar designations. comparison to that in area 24c. To ensure accurate designation of each For abbreviations, see list. Scale bars ϭ 1 mm in top row and bottom layer around the fundus of the sulcus where cortical distortion is row; 0.5 mm in middle row. dorsal and ventral banks are at a higher magnification, former’s extensive numbers of NFP-expressing neurons in although the low magnification at all levels is also quite layer III and much larger Betz neurons in layer Vb (ar- compelling. Area 4mv is also useful in determining the rows in Fig. 6). Differences in superficial layer SMI32 location of layer Vb in SMI32 preparations (Fig. 5). Some immunoreactivity are obvious at all levels of magnifica- differences between areas 4mv and 24d include the tion. Area 24d has a very dense layer Va with both large Fig. 6. Architecture of three levels of midcingulate cortex, includ- also appears that fundal extensions have generally larger neurons in ing the dorsal bank of the cingulate sulcus. SMI32 shows that dorsal layers IIIc and Va. Even at intermediate levels of magnification, it can bank cortex does not share important similarities with cingulate be seen that layer Va neurons are large and there is a progressive cortex; the most profound differences being the very large Betz neu- build-up in small neurons caudally in the sulcus (asterisks). gcc, genu rons in layers Vb of area 4mv (arrows in bottom left plate) and high of the corpus callosum. Scale bars ϭ 0.5 mm (top); 0.5 mm (middle); density of NFPϩ plexi throughout layer V in all cingulate areas. It 250 ␮m (bottom). 228 B.A. VOGT ET AL.

Fig. 7. High magniﬁcations of NeuN in cingulate sulcal areas to intermingled with many small neurons giving the layer a granular assess the composition of layer Va. The asterisks are positioned the appearance. For abbreviations, see list. Scale bars ϭ 500 ␮m (top same way as in Figure 6. Neurons in this layer tend to be larger row); 250 ␮m (bottom row). rostrally. At caudal levels, the neurons can be somewhat smaller and

and small neurons, as is characteristic of most cingulate the primary substrate for motor control as they are areas, and this ﬁnding is not true of area 4mv. “stripped” of all but the largest neurons. The dorsal Figure 1 shows a progressive, rostral-to-caudal build-up midcingulate gyral areas and their fundal extensions are of small neurons in layer Va on the gyral surface of the shown in Figure 8. In all instances, the largest neurons rhesus monkey case and a similar trend occurs on the labeled with SMI32 are larger in the fundal extension. It dorsal bank of the cingulate gyrus in the cynomolgus appears that the largest neurons on the cingulate gyrus monkey. Figure 6 shows that neurons in layer Va of area are in layer Vb of areas fp24cЈ and 24d. Of interest, the a24cЈ are larger and less dense than in area p24cЈ. Area interneurons labeled with parvalbumin are also largest in 24d has a very neuron-dense layer Va with many small layer Va of these same areas. These size differences are neurons intermingled with the larger ones (Fig. 7). Pho- not as noticeable when viewing the total population of tographs at the asterisks were enlarged and rephoto- neurons with NeuN. Finally, the proportionate size differ- graphed for Figure 7 to show the size and density of ences also hold for layer IIIc neurons, which are relatively neurons in all sulcal areas, including area 24c. It appears larger in fundal areas. Thus, the fundal cytoarchitectures that the overall largest neurons are in layer Va of area are not simply the product of laminar and neuronal a24cЈ when viewing NeuN preparations. Although this is stretching, they are also a consequence of pyramidal neu- also true for Nissl-stained tissue, NFP expression by the ron enlargement, and this may be necessary for these largest cingulate neurons shows that these latter are an distorted areas to maintain their functional contribution infrequent part of cingulate architecture and are largest to skeletomotor function. in the fundus adjacent to area 24d. Differentiation of cingulate gyral surface Cingulate sulcus fundal divisions As a general rule, the ventral “a” divisions are relatively Curvature of cingulate cortex around the fundus of the homogeneous when compared with their dorsal “b” coun- cgs greatly attenuates architecture in this region. To the terparts. Because this homogeneity in the former is due to extent that these areas participate in skeletomotor regu- a less well-differentiated layer Va, the “b” divisions also lation, these particular architectures can be considered have a thicker and more-dense layer Va and many more MONKEY CINGULATE GYRUS 229

Fig. 8. The fundal extension of each cingulate gyrus area contains neurons that are generally larger both in SMI32, which emphasizes large, pyramidal-projection neurons, and in parvalbumin, which labels interneurons. “S” is SMI32 and “N” is NeuN. It appears the largest neurons are in layer Vb of area f p24cЈ. For abbreviations, see list. Scale bar ϭ 500 ␮m. 230 B.A. VOGT ET AL.

NFP-expressing neurons in this layer than is the case for areas to show the accurate coregistration in the presence the “a” divisions. Figure 9 provides an intermediate level of an intervening layer IV. This method emphasizes that of magnification of “a:b” pairs of areas at all levels of the the density of NFP-ir neurons is very substantially higher cingulate gyrus. The increase in layer Va neuron density in layer IIIc of v23b and layers IIIc-V are thicker than is and SMI32 immunoreactivity is present at the transition the case for areas d23a or d23b. from area 24a to 24b and a24aЈ to a24bЈ. This difference is Area 31 surrounds the spls and has the most extensive not as pronounced in posterior cortex; however, differ- layers II, IIIab, and IV of any cingulate area. Comparison ences in layer IIIc become more pronounced at caudal of areas 31 and 7m in Figure 11 shows much greater levels (i.e., areas d23a and d23b). NFP-ir in layer IIIc of area 7m, a dense layer IV, and a Level 4 in Figure 9 shows there is no a/b distinction at thinner layer Vb. Higher magnification of layers IIIc–Va the dysgranular area 23d transition. A segment of this shows that neurons throughout mid-cortical layers of area photograph was magnified 2ϫ to show the dysgranular 7m appear to be stacked. The example of one such forma- cortex with an intermingling of large pyramidal neurons tion at the arrow was 37 neurons in length from layer IV in layers IIIc and Va. Also notice that SMI32-ir neurons to its apex in layer IIIc, and its total length was 375 ␮m. are quite similar in all layers along the entire section Of course, neurons form similar aggregates in all cingu- photographed. Although the densities of NFP-expressing late areas; however, those in area 7m are much longer and neurons in layer Va of both divisions of area d23 are contribute in a meaningful way to the low-magnification similar, those in layer IIIc are much higher in the “b” than cytoarchitecture. the “a” division of this area. Another important feature of Architecture of the retrosplenial areas 29 and 30 has cingulate architecture is discernible in Figure 9, where been documented thoroughly in both human and monkey; area 23 is not uniform in its dorsal and ventral parts. Area however, a few comments in the present context including v23b has many more NFP-ir neurons in layers IIIc and Va, parvalbumin expression need consideration. The highest and they are substantially larger than is the case for area level of NFP expression in retrosplenial cortex is in layer d23b as discussed in more detail in the next section. VI and to a lesser extent in layer V of area 29l as shown in Parvalbumin is a sensitive marker of each anterior and Figure 12. The granular neurons of layer III/IV in both midcingulate area as shown in Figure 10. The following areas 29l and 29m do not express these proteins. Because observations are notable: (1) Homogeneity of architecture the laminar position of SMI32 immunoreactivity can be in the dorsal part of area 24a is reflected in parvalbumin difficult to assess, the images were merged with NeuN as immunoreactivity; layer VI has a prominent plexus with shown in Figure 12 to show the exact position of labeled diffuse labeling in layer III of area 24b; and these plexi are neurons. The only NFP immunoreactivity in superficial very prominent, particularly in layer Va in area 24c. (2) layers of area 29 is associated with labeled dendrites from Midcingulate areas a24Ј and p24Ј have very high layer Va deep lying pyramidal neurons. Area 29l has extremely activity, but that in p24Ј is more diffuse. (3) Parvalbumin rich parvalbumin expression, making this marker ideal does not differentiate the posterior cingulate areas partic- for locating the border between 29l and 29m. Indeed, the ularly well, except for the retrosplenial areas as noted labeling of somata and processes is so dense that laminar below. In this regard, area 23d and both divisions of area differentiation is not possible in the former. Finally, area 23 have two plexi in layers III and V that do not vary 30 is dysgranular as shown in human (Vogt et al., 2001). substantially in this region. The variability of layer IV, interdigitation of layer IIIc Posterior cingulate and retrosplenial neurons with those in layer Va, and heavy expression of NFPs by large layer IIIc pyramids is the same as in cortices human. Interestingly, parvalbumin expression is almost Although the posterior cingulate gyrus has been consid- entirely located in layers II–III with a predominance in ered above, the composition of areas 23c, f 23c, 31, and 7m layer II. This pattern of labeling is not present in any have not been addressed nor have the dorsal and ventral anterior or midcingulate area. divisions of PCC been considered in detail. Area 23c on the dorsal bank of the cingulate gyrus (Fig.11; 1) has a broad Horizontal plane of section external granular layer in relation to the deeper layers, Horizontal sections show similar architectures at high dense layers II and IV, extensive NFP-expressing layer magnifications; however, borders between areas on the IIIc pyramids, and a poorly differentiated layer V; all of gyral surface can be identified in macrophotographs of which distinguish it from area 24d rostrally. The fundal single sections and confirm observations of different ar- division of area 23c is similar to its dorsal bank counter- chitectures in separate coronal sections. Figure 13 shows part, although the lamination is greatly attenuated. a macrophotograph through the dorsal part of the cingu- Areas 23a/b are not uniform in the dorsoventral plane or late gyrus with progressive differentiation from rostral “z” axis. The dorsal areas d23a/b are relatively thinner area 24b to caudal area 31. Layer Va in the NeuN sample than their ventral counterparts and have a slightly more- (Fig. 13A) is emphasized with arrows to show the progres- dense layer Va. As shown in Figure 9 (NeuN), area v23b sion from thin and relatively sparse in area 24b (1) to a has much thicker layers IIIc, IV, and V, and it has many thicker and more neuron-dense composition particularly more NFP-ir neurons (SMI32) in layers IIIc and Va, and in areas p24bЈ, 23d, and d23b (3, 4). Progressive changes they are substantially larger than is the case for area in the expression of neurofilament proteins are also shown d23b. The NeuN and SMI32 images were merged by re- Figure 13B, where arrows are used to emphasize changes ducing the opacity of the SMI32 image and magnifying in layer IIIc. Before arrow 5 in Figure 13, the SMI32 both by 1.5ϫ to show the exact position of the SMI32-ir immunoreactive neurons are sparse in layer IIIc and tend neurons in both areas. Although all pairs of NeuN and to be solitary. The border of areas p24bЈ and 23d is her- SMI32 images were so coregistered to ensure exact coreg- alded by heavy expression of NFP by layer IIIc pyramidal istration of images, the present ones are provided for these neurons starting at arrow 5. The sixth arrow emphasizes MONKEY CINGULATE GYRUS 231

Fig. 9. Architecture and cytology of cortex along the surface of the cortex with the interdigitation of layer IIIc and Va pyramids. Differ- cingulate gyrus. Comparisons of the “a” and “b” divisions are en- ences between the dorsal and ventral parts of area 23 are shown with hanced with the vertical arrows to the border of layers IIIc and Va. levels 5 and 6. To ensure exact coregistration of the layers, the opacity Dysgranular area 23d has a variable layer IV and relatively even of the SMI32 photographs was reduced and the NeuN image merged distribution of NFP-immunoreactive neurons in layers IIIc and Va. with it at increased magniﬁcation (1.5ϫ). For abbreviations, see list. Layer IV is magniﬁed (2ϫ) to show the dysgranular nature of this Scale bar ϭ 500 ␮m. 232 B.A. VOGT ET AL.

Fig. 10. Parvalbumin-immunoreactivity is helpful for identifying area 24 at the genu of the corpus callosum. Even differences between areas in anterior and midcingulate cortices, whereas it is less valuable a24Ј and p24Ј can be seen with the broader layer Va and more diffuse with areas 23 and 31. For example, level 1 shows the four divisions of plexi in the latter areas. For abbreviations, see list. Scale bar ϭ 1 mm. the intermingling of NFP-expressing neurons in layer IIIc DISCUSSION with those in layer Va in dysgranular area 23d. Area d23b has a more-dense expression of neurons in both layers IIIc The present analysis of monkey cingulate gyrus archi- and Va, whereas those in both layers of area 31 are among tecture and neurocytology builds upon a rich base of hu- the highest in the cingulate gyrus. Each of these laminar man anatomical and functional studies, and it appears observations can be verified above with high magnification that the fundamental organization of monkey cingulate and in the coronal plane of section. cortex is similar to that in human. Particular new findings MONKEY CINGULATE GYRUS 233

Fig. 11. Dorsal parts of PCC (1) at area 23c, 2 at area 31, and 3 at 31, very large and dense, NFP-expressing, layer IIIc pyramids, and medial parietal area 7m. Area 23c has a disproportionately broad small neurons that extend into layer IIIc. The stacking of neurons in superﬁcial layer and compressed layer V. Area 31 has the thickest mid-cortical layers is characteristic of area 7m (asterisks; magniﬁca- layer IV of any cingulate area and high densities of layer IIIc NFP- tion 2ϫ) and not of area 31. For abbreviations, see list. Scale bar ϭ expressing neurons. Area 7m has a much broader layer IV than area 250 ␮m. 234 B.A. VOGT ET AL.

Fig. 12. Retrosplenial areas are well deﬁned with NeuN, SMI32, from NeuN and SMI32. Area 30 is dysgranular as seen with NeuN, and parvalbumin. Indeed, the massive intrinsic parvalbumin system and the highest level of NFP expression is in layers IIIc and Va as is very robust in area 29l. Neurons in the granular layer (i.e., layer emphasized in the merged sections. For abbreviations, see list. Scale III/IV) do not express NFP, and this is documented by merging images bar ϭ 500 ␮m. in monkey include (1) two divisions of MCC (a24Ј and dorsal bank of the cingulate sulcus; (4) parvalbumin is a p24Ј); (2) two divisions of PCC (d23 and v23); (3) fundal useful marker of areas in ACC and RSC, although less extensions of all areas on the dorsal bank of the cingulate useful for divisions of PCC; and (5) area 23d is a dysgranu- gyrus, however, no cingulate areas are formed on the lar transition area between areas p24aЈ/bЈ and d23a/b. MONKEY CINGULATE GYRUS 235

Fig. 13. The dorsal part of the cingulate gyrus shows progressive neurons in layer IIIc to the border of areas p24bЈ and 23d with heavy differentiation from area 24b to area 31. The asterisk on the medial expression of NFP by layer IIIc pyramidal neurons (5). Area d23b has surface orients to the same dimple in each section. A: Layer Va in a more-dense expression of neurons in both layers IIIc and Va, NeuN is emphasized with arrows to show progression from thin and whereas those in both layers of area 31 are among the highest density relatively sparse (1) to a thicker and more neuron-dense composition and intensity in the cingulate gyrus (8). For abbreviations, see list. (3, 4). B: Progressive changes in the expression of NFP is shown, Scale bar ϭ 1 mm/division (above), 1 mm (below). where arrows emphasize changes in layer IIIc from sparse SMI32ϩ

Flat maps of area distributions on the cingulate gyrus can sion, this region has altered neuronal activity and may be used for direct comparisons of connection and motor explain mood-congruent processing disruption with re- system organization. duced responses to happiness when compared with controls and elevated responses during sadness (Elliott et al., Anterior cingulate region 2002). Of interest, sadness activity occurred ventral to the The four-region model of the cingulate gyrus places ACC genu in healthy individuals, whereas happiness was more in a perigenual position in human (Vogt et al., 2004) and rostral (Vogt et al., 2003). monkey (Vogt et al., 1997), and this and other changes have It is important that another key rationale for the ACC/ been introduced in the ACC since the original map was MCC dissociation is the selective vulnerability of ACC to published in 1987. First, area 24a in the monkey appears to major depression. Glucose metabolism is reduced in this be the major area caudal to area 32 with little or no contri- region in major depression (Drevets, 2001); however, dif- bution of area 24b as suggested by Carmichael and Price ferences in glucose metabolism for drug responders and (1994). These latter investigators also demonstrated a me- nonresponders occurs dorsal in it (Mayberg et al., 1997). dial extension of area 10 termed area 10m ventral to area 32, At present, the entire ACC must be viewed as vulnerable and this finding was confirmed in the present study with all to depression, whereas MCC appears to have no such three markers. Braak (1979) evaluated the perigenual re- vulnerability. Although there is conflicting evidence for a gion in human pigment architecture preparations and re- structural basis for depression (Harrison, 2002), the orga- ferred to it as a magnocellular area. Review of the structure nization, connections, and chemistry of this region in mon- of area 24a in this region in monkey suggests that it does not key and its potential to develop a model of depression have disproportionately large neurons. Rather, the largest makes further assessment of ACC pivotal to understand- neurons lie in each fundal division of the dorsal bank of the ing human neuronal diseases. cingulate gyrus, including those in area f 24c. The role of ACC in autonomic regulation is well-known Midcingulate cortex and serves as an important criterion for defining this The most profound change to the concept of monkey region as separate from MCC (Neafsey et al., 1993; Vogt et cingulate cortex deriving from the present work is that al., 1997). The subgenual part of ACC is activated during MCC is a unique division. Although first proposed as a sad events in healthy human subjects (Mayberg, 1997; transitional region in monkey (Vogt, 1993), this view Phan et al., 2002; Vogt et al., 2003), and in major depres- needs to be extended as in human. Area 24Ј is more than 236 B.A. VOGT ET AL. a structural transition cortex, because it represents a al. (2004), although it may have some corticospinal pro- unique structural region with specific functions that dif- jection neurons. ferentiate it from the other regions. Moreover, it composes Figure 14 provides a comparison of current findings fully one third of the entire cingulate gyrus. Thus, area 24Ј with those of Dum and Strick (1991). Although the coreg- represents more than a simple transitional region; it is a istration is approximate, it appears that corticospinal pro- unique region, and the MCC conveys this intention rather jections originate from area 23c, even though stimulation- than the view of a transition area. Two points about cin- evoked movements were not observed by Rizzolatti et al. gulate regionalization of structure and function flow from (1996). The greatest density of corticospinal projection this perspective. First, transition between ACC and PCC neurons is in areas p24cЈ and fp24cЈ; the latter of which we is not due to a simple deposition of more neurons in layer observed to have the largest, layer-Vb, pyramidal neu- IV as suggested by Brodmann (1909), Vogt et al. (1987), rons. Finally, corticospinal projections are prominent from and Morecraft et al. (2004). According to further com- area a24cЈ and very limited from area 24c. The two cingu- ments below, there is a progression that actually starts late motor areas likely reside in areas 24d/p24cЈ (cCMA) with differentiation of layer Va before addition of a dys- and area a24cЈ (rCMA). granular area 23 and the granular areas 23 and 31. Sec- It has been proposed that there is a third CMA on the ond, the concept of MCC emphasizes its unique role as a dorsal bank of the caudal cingulate sulcus termed area 6c skeletomotor control region with specific architectures, or “dCMA” (Dum and Strick, 1993; He et al., 1995). No projection neurons, and functional responses, including cortex dorsal to the fundal divisions has a “cingulate” nociceptive ones. motif (e.g., large and dense layer Va relative to layer IIIc In view of the ACC/MCC differentiation in the monkey, in NeuN, and heavier NFP immunoreactivity in layer Va it is possible to use the monkey to model particular neu- than layer IIIc). It appears that cortex on the dorsal bank ronal diseases. The MCC in human is selectively vulner- of the cgs is composed of ventral divisions of areas 6a␤ and able to chronic neuropathic and lower-back pain syn- 6a␣ (or area 6c according to Dum and Strick and He et al.) dromes (Hsieh et al., 1995; Derbyshire et al., 1994, 2002), rather than cingulate cortex as observed in human brain whereas syndromes associated with psychiatric symptoms (Vogt and Vogt, 2003). Therefore, this dorsal cortex is selectively impact ACC as discussed above. Thus, this either part of the supplementary motor system or it is a differential vulnerability to neuronal disease is a key cri- unique extension thereof. Russo et al. (2002) attempted to terion for differentiating ACC and MCC, and their pres- evaluate the “dCMA” and suggested similarities between ence in monkey is yet another tool for studying human it and the cCMA. Although no medial surface landmarks disease processes. are provided, it appears that 13 of their 18 presumptive Cingulate motor areas “dCMA” units were in the fundal part of area p24cЈ rather than on the dorsal bank of the cgs per se; only two units Braak (1976) first demonstrated a CMA with his prim- were on the dorsal bank. Thus, there is not yet adequate itive gigantopyramidal field. A similar area in monkey evidence for a third CMA in the dorsal bank of the caudal was identified and termed area 24d (Matelli et al., 1991); cingulate sulcus, and it will likely not be a “cingulate” it was shown to have large and NFP-ir neurons in layer Vb (Nimchinsky et al., 1996), and it has a topographically motor area if it is found. This conclusion raises further organized motor field (Rizollatti et al., 1996). The organi- issues, however, particularly in terms of whether or not zation of sulcal cortex, however, is quite complex, and the there is a single somatotopic map that extends across the anatomical composition and functions of this region are dorsal and ventral banks of the cgs. Labeling of cortical still a matter of active research and debate. Zilles et al. projection neurons and electrical stimulation suggests a (1996) observed that the caudal cingulate motor area in single map in the cCMA that may extend beyond cingulate human (cCMA; area 24d) is composed of two divisions; cortex (Wang et al., 2001). This explanation suggests that cmc1 and cmc2. Area cmc1 is caudal to the latter and has cortical and intrinsic processing must be substantially much larger neurons in layers III and V. The present different for regions involving forelimb and upper trunk monkey observations are in agreement with those of Zilles versus hindlimb and lower trunk projection cortex. The et al. (1995; 1996) in the human. Thus, area cmc1 equates extremely dense NFP-expressing neurons in layer IIIc of ␤ to area 24d and cmc2 to area p24cЈ. areas 6c/v6a suggest a very different phenotype for each To evaluate the projections and electrically stimulated bank of the cgs. movements in the context of the current observations, Although the organization of cingulate motor cortex is Figure 14 shows the flat map coregistered to findings of becoming quite clear, it does appear that each area has Rizzolatti et al. (1996) and Dum and Strick (1991). Al- substantial intrinsic variability. What is the interpreta- though the overlap is only approximate, several interest- tion of widely variable cytoarchitectures in a single, soma- ing suggestions emerge from comparing these various re- totopically organized motor field? Although it is true that ports. Rizzolatti et al. (1996) extended area 24d rostrally corticospinal projections provide a unifying feature of the to include much of our area p24cЈ. Much of area 24d is cingulate motor areas (Dum and Strick, 1991, 1993), involved in forelimb, upper trunk, and neck movement, structural differences within a motor field suggest they whereas area p24cЈ appears to be involved in hindlimb, have different intrinsic organization and contribute to lower trunk, and tail movements. Notice that Matelli et al. additional functions. In other words, cortex of the cCMA (1991) included a rostral extension of our fp24cЈ in their may contribute to cognitive functions beyond simple skel- area 24d. Of interest, area fp24cЈ has the largest neurons etomotor regulation. The present findings suggest that the in the cingulate gyrus. We fully agree with the localization monkey cCMA does not have an extension onto the dorsal of area 23c by Matelli et al. (1991) and notice that it had bank of the cingulate sulcus as in human (Vogt et al., no evoked activity (Rizzolatti et al. (1996). Therefore, area 2004) and that cortex subsuming the cCMA includes areas 23c may not contain a CMA as suggested by Morecraft et 24d and p24cЈ, which may equate to cmc1 and cmc2, MONKEY CINGULATE GYRUS 237

Fig. 14. The flat map coregistered to the findings of other investi- evoked forelimb–upper trunk–neck movements. Nonresponsive sites gators. A,B: In view of the pivotal nature of corticospinal projections are not shown for ease of comparison. B: The density of labeled in the cingulate sulcus, electrical stimulation (A, modified from Riz- neurons is shown after a horseradish peroxidase injection into the zolatti et al., 1996) and labeling of corticospinal projection neurons (B, cervical spinal cord to produce a relatively complete labeling of corti- modified from Dum and Strick, 1991) were coregistered to the flat cospinal projection neurons. In both instances, area designations are map. In both cases, the dashed line represents the fundus of the shown to correlate with matches and mismatches between our find- cingulate sulcus. A: The open circles represent sites that activated ings and previous observations in the cingulate sulcus. For abbrevi- hindlimb–lower trunk–tail movements, whereas the filled circles ations, see list.

respectively, of Zilles et al. (1996). There also are the two ventral bank counterparts. Anatomical variations in the fundal extensions of each of these areas, which always cingulate motor ﬁelds may lead to a new view of cognitive– contain larger neurons in layers IIIc and Vb than do their motor interactions (Murtha et al., 1996; Bush et al., 2002). 238 B.A. VOGT ET AL.

Because the fundal extensions of the CMAs have pro- dPCC receiving inputs from the central laterocellular, me- foundly distorted architectures due to cortical curvature, diodorsal, and ventral anterior and ventral lateral nuclei it is unlikely that they contribute equally to function like that do not project to the vPCC. Finally, neurocytology in other parts of the motor fields. Indeed, in addition to the human cortex using techniques similar to those of the having the largest pyramidal neurons in the cingulate present study also showed a cytological and functional gyrus, the density and size of parvalbumin-expressing difference between these areas (Vogt et al., personal com- neurons is substantially greater in fundal divisions. Be- munication). cause fundal areas 24d and p24cЈ participate in skeleto- The morphological substrate for defining the two parts motor regulation, it might be posited that cingulate motor of area 23 include the many more and larger neurons in function is based on the organization of the fundal regions layers IIIc and V, more NFP-expressing neurons in these and that additional architectures encountered on the dor- layers, and the much thicker layer IV in area v23 than in sal bank of the cingulate gyrus represent additional cir- d23. These are the same features that distinguish these cuitry for cognitive processing. The substantial variation areas in human (Vogt et al., personal communication). in structural organization even within a single cingulate The vPCC region has direct and reciprocal interactions motor area predicts new functions for these areas, which with subgenual ACC (Vogt and Pandya, 1987), and this extend beyond simple motor output. linkage provides for direct interchanges of information Dysgranular cortex and cingulate between these two regions. Because the former is involved in affect and emotion, as discussed in the introduction, transitions and the latter is involved in visuospatial orientation, it is Area 30 is a dysgranular transitional cortex between possible that information interchanges by means of the areas 29m and 23a that has heavy and reciprocal connec- vPCC are associated with the personal relevance of previ- tions with area 23a (Vogt and Pandya, 1987). Because the ously coded sensory events. present study also demonstrated a dysgranular area 23d between areas p24aЈ/bЈ and d23a/b that is continuous with area 30 of RSC, there appears to be a dysgranular tran- LITERATURE CITED sition zone between the agranular and periallocortical regions of the cingulate gyrus. One of the most important Akkal D, Audin J, Burbaud P. 2002. Comparison of neuronal activity in the rostral supplementary and cingulate motor areas during a task with issues raised in the past decade of human cytological cognitive and motor demands. Eur J Neurosci 15:887–904. research is the location of the border between anterior Baleydier C, Mauguiere F. 1980. The duality of the cingulate gyrus in agranular area 24 and posterior granular area 23. The monkey: neuroanatomical study and functional hypothesis. Brain 103: Brodmann map shows a rather simple and sharp transi- 525–554. tion from an agranular area 24 to a granular area 23, and Biber MP, Kneisley LW, LaVail JH. 1978. Cortical neurons projecting to this is reflected in our earlier map (Vogt et al., 1987); a the cervical and lumbar enlargements of the spinal cord in young and thorough re-analysis of this case does not show such a adult rhesus monkeys. Exp Neurol 59:492–508. Braak H. 1976. A primitive gigantopyramidal field buried in the depth of simple transition. In the context of pMCC, the transition the cingulate sulcus of the human brain. Brain Res 109:219–233. between these two fundamentally different types of cortex Braak H. 1979. Pigment architecture of the human telencephalic cortex. V. can be detailed: (1) Between areas a24Ј and p24Ј layer Va Regio anterogenualis. Cell Tissue Res 204:441–451. has a substantial increase in the number of small neurons Broca P. 1878. Anatomie compareé circonvolutions ce´re´brales: Le grande in layer Va. (2) Between areas p24Ј and 23d, there is an lobe limbique et la scissure limbique dans la se´rie des mammife`res. Rev additional dysgranular layer IV. (3) Between areas 23d Anthropol Ser 21:384–498. and 23a/b, there is an increase in the breadth of layer IV. Brodmann K. 1909. Vergleichende Lokalisationslehre der Grosshirnrinde in ihren Prinzipien dargestellt auf Grund des Zellenbaues. Leipzig: Of interest, even in granular area 23 in NeuN prepara- Barth. tions, layer Va is more prominent than is layer IV, because Bush G, Vogt BA, Holmes J, Dale AM, Greve D, Jenike MA, Rosen BR. of the granularity of the former. (4) Between areas d23a/b 2002. Dorsal anterior cingulate cortex: a role in reward-based decision- and 31, there is a profound increase in the breadth of layer making. Proc Natl Acad SciUSA99:523–528. IV and a shift in larger pyramids to layer IIIc from layer Carmichael ST, Price JL. 1994. Architectonic subdivision of the orbital and Va. In addition to this latter change, the density of NFP- medial prefrontal cortex in the macaque monkey. J Comp Neurol expressing neurons in layer IIIc also increases substan- 346:366–402. Derbyshire SWG, Jones AKP, Devani P, Friston KJ, Feinman C, Harris M, tially. Thus, the Brodmann map misidentifies the point of Watson JD, Frakowiak RSJ. 1994. Cerebral responses to pain in pa- transition from agranular area 24 to granular area23, and tients with atypical facial pain measured by positron emission tomog- the complex features of transition between anterior and raphy. J Neurol Neurosurg Psychiatry 57:1166–1172. posterior cingulate cortices have been thoroughly defined Derbyshire SWG, Jones AKP, Creed F, Starz T, Meltzer CC, Townsend for both monkey and human brains. DW, Peterson AM, Firestone L. 2002. Cerebral responses to noxious thermal stimulation in chronic low back pain patients and normal Dorsal and ventral PCC controls. Neuroimage 16:158–168. Dombrowski SM, Hilgetag CC, Barbas H. 2001. Quantitative architecture Circuitry and functional studies show differentiation of distinguishes prefrontal cortical systems in the rhesus monkey. Cereb PCC into two parts, yet no cytoarchitectural studies have Cortex 11:975–988. analyzed whether or not there is a structural substrate for Drevets WC. 2001. Neuroimaging and neuropathological studies of depres- it. Thus, auditory inputs arrive to dPCC from caudal area sion: implications for the cognitive-emotional features of mood disor- TA and those to vPCC from rostral area TA (Yukie, 1995). ders. Curr Opin Neurobiol 11:240–249. Prefrontal cortex projections from the dorsal bank of the Dum RP, Strick PL. 1991. The origin of corticospinal projections from the premotor areas in the frontal lobe. J Neurosci 11:667–689. principal sulcus terminate in dPCC, whereas cortex in the Dum RP, Strick PL. 1993. Cingulate motor areas. In: Vogt BA, Gabriel M, rostral tip of both banks projects to vPCC (Vogt and Bar- editors. Neurobiology of cingulate cortex and limbic thalamus. Boston: bas, 1988). Shibata and Yukie (2003) reported that each Birkhauser. p 415–441. PCC division has unique thalamic connections with the Elliott R, Rubinsztein JS, Sahakian BJ, Dolan RJ. 2002. The neural basis MONKEY CINGULATE GYRUS 239

of mood-congruent processing biases in depression. Arch Gen Psychia- signal-triggered and self-paced forelimb movements. J Neurophysiol try 59:597–604. 65:188–202. Harrison PJ. 2002. The neuropathology of primary mood disorder. Brain Smith GE. 1907. A new topographical survey of the human cerebral cortex, 125:1428–1449. being an account of the distribution of the anatomically distinct cortical He S-Q, Dum RP, Strick PL. 1995. Topographic organization of corticospi- areas and their relationship to the cerebral sulci. J Anat 41:237–254. nal projections from the frontal lobe: motor areas on the medial surface Takada M, Tokuno H, Hamada I, Inase M, Ito Y, Imanishi M, Hasegawa N, of the hemisphere. J Neurosci 15:3284–3306. Akazawa T, Hatanka N, Nambu A. 2001. Organization of inputs from Hsieh J-C, Belfrage M, Stone-Elander S, Hansson P, Ingvar M. 1995. cingulate motor areas to basal ganglia in macaque monkey. Eur J Neu- Central representation of chronic on going neuropathic pain studied rosci 14:1633–1650. with positron emission tomography. Pain 63:225–236. Van Hoesen GW, Morecraft RJ, Vogt BA. 1993. Connections of the monkey Luppino G, Matelli M, Camarda RM, Gallese V, Rizzolatti G. 1991. Mul- cingulate cortex. In: Vogt BA, Gabriel M, editors. Neurobiology of tiple representations of body movements in mesial area 6 and the cingulate cortex and limbic thalamus. Boston: Birkhauser. p 249–284. adjacent cingulate cortex: An intracortical microstimulation study in Vogt BA. 1993. Structural organization of cingulate cortex: areas, neurons, the macaque monkey. J Comp Neurol 311:463–482. and somatodendritic transmitter receptors. In: Vogt BA, Gabriel M, MacLean PD. 1990. The triune brain in evolution, role in paleocerebral editors. Neurobiology of cingulate cortex and limbic. Boston: functions. New York: Plenum Press. Birkhauser. p 19–70. Matelli M, Luppino G, Rizzolatti G. 1991. Architecture of superior and Vogt B, Barbas H. 1988. Structure and connections of the cingulate vocalization region in the rhesus monkey. In: Newman JD, editor. The mesial Area 6 and the adjacent cingulate cortex in the macaque mon- physiological control of mammalian vocalization. New York: Plenum key. J Comp Neurol 311:445–462. Press. p 203–225. Mayberg HS. 1997. Limbic-cortical dysregulation: a proposed model of Vogt BA, Pandya DN. 1987. Cingulate cortex of the rhesus monkey: II. depression. J Neuropsychiatry 9:471–480. Cortical afferents. J Comp Neurol 262:271–289. Mayberg HS, Brannan SK, Mahurin RK, Jerabek PA, Tekell JL, McGinnis Vogt BA, Vogt L. 2003. Cytology of human dorsal midcingulate and sup- S, Glass TG, Martin CC, Fox PT. 1997. Cingulate function in depres- plementary motor cortices. J Chem Neuroanat 26:301–309. sion: a potential predictor of treatment response. Neuroreport 8:1057– 1061. Vogt BA, Rosene DL, Pandya DN. 1987. Cingulate cortex of the rhesus monkey: I. Cytoarchitecture and thalamic afferents. J Comp Neurol Morecraft RJ, Van Hoesen GW. 1992. Cingulate input to the primary and 262;256–270. supplementary motor cortices in the rhesus monkey: evidence for so- matotopy in Areas 24c and 23c. J Comp Neurol 322:471–489. Vogt BA, Finch DM, Olson CR. 1992. Functional heterogeneity of cingulate cortex: The anterior executive and posterior evaluative regions. Cereb Morecraft RJ, Cipollomi PB, Stilwell-Morecraft KS, Gedney MT, Pandya Cortex 2:435–443. DN. 2004. Cytoarchitecture and cortical connections of the posterior cingulate and adjacent somatosensory fields in the rhesus monkey. Vogt BA, Nimchinsky EA, Vogt LJ, Hof PR. 1995. Human cingulate cortex: J Comp Neurol 469:37–69. surface features, flat maps, and cytoarchitecture. J Comp Neurol 359: 490–506. Murtha S, Chertkow H, Beauregard M, Dixon R, Evans A. 1996. Anticipa- tion causes increased blood flow to the anterior cingulate cortex. Hum Vogt BA, Vogt LJ, Nimchinsky EA, Hof PR. 1997. Primate cingulate cortex Brain Map 4:103–112. chemoarchitecture and its disruption in Alzheimer’s disease. In: Bloom FE, Bjorklund A, Hokfelt T, editors. Handbook of chemical neuroanat- Neafsy EJ, Terreberry RR, Hurley KM, Ruit KG, Frysztak RJ. 1993. omy, the primate nervous system. Vol. 13. Part I. Amsterdam: Elsevier. Anterior cingulate cortex in rodents: connections, visceral control func- p 455–528. tions, and implications for emotion. In: Vogt BA, Gabriel M, editors. Neurobiology of cingulate cortex and limbic thalamus. Boston: Vogt BA, Berger GR, Derbyshire SWG. 2003. Structural and functional dichotomy of human midcingulate cortex. Eur J Neurosci 18:3134– Birkhauser. p 206–223. 3144. Nimchinsky EA, Hof PR, Young WG, Morrison JH. 1996. Neurochemical, Vogt BA, Hof PR, Vogt L. 2004. Cingulate gyrus. In: Paxinos G, Mai J, morphologic, and laminar characterization of cortical projection neu- editors. The human nervous system. 2nd ed. San Diego: Elsevier. p rons in the cingulate motor areas of the macaque monkey. J Comp 915–945. Neurol 374:136–160. Vogt BA, Vogt LJ, Perl DP, Hof PR. 2001. Cytology of human caudomedial Nimchinsky EA, Vogt BA, Morrison JH, Hof PR. 1997. Neurofilament and cingulate, retrosplenial, and caudal parahippocampal cortices. J Comp calcium-binding proteins in the human cingulate cortices. J Comp Neurol 438:353–376. Neurol 384:597–620. Vogt C, Vogt O. 1919. Allgemeinere Ergebnisse unserer Hirnforschung. Papez JW. 1937. A proposed mechanism of emotion. Arch Neurol Psychi- J Psychol Neurol 25:279–461. atry 38:725–733. von Economo C, Koskinas GN. 1925. Die Cytoarchitectonik der Hirnrinde Phan KL, Wager T, Taylor SF, Liberzon I. 2002. Functional neuroanatomy des erwachsenen Menschen. Berlin: Springer. of emotion: a meta-analysis of emotion activation in PET and fMRI. Neuroimage 16:331–348. Vorobiev V, Govoni P, Rizzolatti G, Matelli M, Luppino G. 1997. Parcella- tion of human mesial area 6: cytoarchitectonic evidence for three sep- Rizzolatti G, Luppino G, Matelli M. 1996. The classic supplementary motor arate areas. Eur J Neurosci 10:2199–2203. area ids formed by two independent areas. Adv Neurol 70:45–56. Wang Y, Shima K, Sawamura H, Tanji J. 2001. Spatial distribution of Rose M. 1927. Gyrus limbicus anterior and Regio retrosplenialis (Cortex cingulate cells projecting to the primary supplementary and pre- holoprotoptychos quinquestratificatus) Vergleichende Architektonik supplementary motor areas: a retrograde multiple labeling study in the bei Tier und mensch. J Psycol Neurol 43:65–173. macaque monkey. Neurosci Res 39:39–49. Russo GS, Backus DA, Ye S, Crutcher MD. 2002. Neural activity in mon- Yukie M. 1995. Neural connections of auditory association cortex with the key dorsal and ventral cingulate motor areas: Comparison with the posterior cingulate cortex in the monkey. Neurosci Res 22:179–187. supplementary motor area. J Neurophysiol 88:2612–2629. Zilles K, Schlaug G, Matelli M, Luppino G, Schleicher A, Qu M, Dabring- Shibata H, Yukie M. 2003. Differential thalamic connections of the pos- haus A, Seitz R, Roland PE. 1995. Mapping of human and macaque teroventral and dorsal posterior cingulate gyrus in the monkey. Eur sensorimotor areas by integrating architectonic, transmitter receptor, J Neurosci 18:1615–1626. MRI and PET data. J Anat 187:515–537. Shima K, Tanji J. 1998. Role for cingulate motor area cells in voluntary Zilles K, Schlaug G, Geyer S, Luppino G, Matelli M, Qu¨ M, Schleicher A, movement selection based on reward. Science 282:1335–1338. Schormann T. 1996. Anatomy and transmitter receptors of the supple- Shima K, Aya K, Mushiake H, Inase M, Aizawa H, Tanji J. 1991. Two mentary motor areas in the human and nonhuman primate brain. Adv movement-related foci in the primate cingulate cortex observed in Neurol 70:29–43.